Wan Open Models: Model List and Licenses

September 27, 2026

About Wan

Wan is a series of open video generation models. Its Wan-AI organization on Hugging Face hosts the video generation models from the paper “Wan: Open and Advanced Large-Scale Video Generative Models". It is one of the most widely used families of video generation models you can run locally.

What It Releases

• Wan2.2 (July 2025): according to its model card, its key feature is bringing an MoE architecture into video diffusion models, splitting the denoising process across timesteps among specialized expert models to enlarge capacity without raising compute. Curated aesthetic data with detailed labels for lighting, composition and color tone gives finer control over cinematic style, and it was trained on 65.6% more images and 83.2% more videos than Wan2.1.
• T2V-A14B and I2V-A14B: MoE models for text-to-video and image-to-video (480P and 720P) – TI2V-5B: a 5B model with its own high-compression VAE (16×16×4) that makes 720P, 24fps video from text or images. The model card says it runs on a single consumer GPU such as the RTX 4090.
• Wan2.2 derivatives: S2V, which makes video from audio (August 2025), and Animate for character animation (September 2025 and August 2026).
• Wan-Dancer (July 2026): generates minute-scale dance videos synchronized with music.

Licenses

All the models we checked are Apache-2.0, broadly usable including commercially. Read each model’s license text before use. This summary is not legal advice.

Running Them Locally

• On a consumer GPU, TI2V-5B is the easiest option. The A14B models need quantization or loading components in parts.
• There are no official GGUF builds (as of 2026-09-26). GGUF and FP8 builds for ComfyUI come from third parties; official diffusers versions are available.
• Video models load a text encoder and VAE besides the main model. Check the memory needed for all components in the tables in our articles.

Sources: model cards of Wan-AI/Wan2.2-T2V-A14B, Wan-AI/Wan2.2-TI2V-5B and Wan-AI/Wan-Dancer-14B; the model list of the Wan-AI organization on Hugging Face (all as of 2026-09-26).

Our Coverage and Data

Local Model Watch has not published an article on Wan’s models yet. The model list below is collected by code from Hugging Face every day. Part of our publisher index; model families link to their family pages.

At a Glance

Recent Models on Hugging Face

The publisher lists 27 models in total on Hugging Face. Below are up to 15 of the most recent generative models (text, image, video, audio and so on), the scope of this site, per organization. Classifiers, feature extractors, research components and repositories without a model card are omitted, and the publisher’s own quantized builds and helper variants are folded into the row of the original model. Parameter counts are left blank for repositories that pack weights into integer types, where Hugging Face’s count is not the real parameter count.

From the Hugging Face API. Last changed 2026-09-27 (JST). Collected by code.

Last updated 2026-09-27 (JST). The explanation at the top of this page was written with the help of AI from the primary sources it cites. The tables and lists under “Our Coverage and Data" are assembled by code from our article log and the Hugging Face API.